arrow_back Back to round 8

Chess Olympiad Samarkand 2026 · Round 8 · Board 2

Georgiadis vs Bakhrillaev

Loading game

Live broadcast by lichess.org

GM Nico Georgiadis (SUI, 2535) vs IM Bakhrom Bakhrillaev (UZB, 2412) · ½-½

Board 2, Switzerland vs Uzbekistan 3 · Queen's Gambit Declined, Exchange Variation, Carlsen Variation (D35) · 90 minutes for 40 moves, then 30 minutes, plus 30 seconds per move

By Professor Zugzwang AI Coach . This analysis was written by a Bogner Chess AI coach.

Round eight, board two, White against an International Master rated 2412, a good 120 points below Nico Georgiadis on paper. On board one Sebastian Bogner had just beaten Saidakbar Saydaliev with Black, and what the team needs on a day against Uzbekistan 3 is a second point on one of the top boards, not a half. Georgiadis met the Queen's Gambit Declined with the Exchange Variation and , the quiet version, in which White plays on space and not on tactics.

For twenty moves the plan worked, and by move thirty-four the position stood a pawn and a half better for White. What it did not survive was the clock: as in round six, less than three minutes were left for the moves that would have brought the point home. On move thirty-five the one move that gave the position away was played, and on moves thirty-eight and forty the one that would have won it back was missing.

The square behind the knight: 22.Bxe4

The knight on e4 is a nuisance: it cuts the rook on f3 off from its diagonal, it attacks the knight on c3, and it gives Black an anchor in the middle of the white position. clears it out and leaves two things behind in the trade that fit together exactly. One is the pawn on e4, which blocks the h1 diagonal and takes care of the fact that White's light-squared bishop is now off the board. The other is the d5 square, no longer occupied by the d-pawn: a hole around which White builds the game for the next ten moves.

The plan behind the move is the kingside push. fixes the pawns, Black on f5 and White on g5, and the black king position is now locked shut on the dark squares: no black pawn ever comes past g5, and around g5 the white knight on h4 can drop into f5 whenever it wants to. and then arranged the white pieces so that the queens came off with without a lost tempo, and by move thirty-four the assessment stood at about a pawn and a half in White's favour.

Before you give a bishop for a knight, check not what the bishop gives up, but what the square behind the knight does after the trade. A knight that comes off is a knight whose defenders lose their grip, and a pawn that recaptures is a pawn that vacates the square behind it. In fixed pawn structures that one square decides the rest of the game.

IM Bakhrom Bakhrillaev (UZB, 2412)

GM Nico Georgiadis (SUI, 2535)

After 21...Ne4: the knight sits in the middle of White's set-up, defended by the pawn on d5. White to move.

The move in forty-eight seconds: 35.b4?

The position looks quiet and is not. The c4 pawn hangs twice and is defended twice, the pawn chain on b3 and d4 holds the structure together, and the one move that turns a plus of about a pawn and a half into a win is the break: , and White has a clean passed pawn on a2, the bishop pair no longer matters, and the h-file opens. Almost half of all players go bxc4 here, and the move is the practically easy one too.

? does the opposite. It closes the queenside and freezes the structure that White has been working on: the pawns on b3, b4 and d4 become a wall, the black pawn on c4 is now passed rather than attacked, and the d5 square, on which the whole game rested, is free for the knight on b6. After the position stood objectively at 0.13, and that was the end of the advantage. Not a calculation error in this game, a tactical one: the only move that carried a winning try was the pawn trade in the middle, and the move that avoided it gave the rest away with it.

Before you close the wing, count what the break gives you. The pawn move that removes tension is never neutral in a better position: it is either the way you consolidate an advantage or the way you lose one. Which category it falls into is decided not by how the move looks, but by what happens along the file you are closing.

IM Bakhrom Bakhrillaev (UZB, 2412)

GM Nico Georgiadis (SUI, 2535)

After 34...Rdd8: the c4 pawn is attacked twice and defended twice. White to move, with one minute and thirty-four seconds on the clock.

The wrong answer twice: 38.Ne2? and 40.Ne2?

After ? Black should hold the position with , but he swapped that knight for a manoeuvre meant to reach the same position three times: leaves the strong square on d5 and lures the white knight away from f4. The only move that wins the advantage back is , and it is also best: , and White stands a pawn to the good and clearly better again, with the b5 pawn taken and an h4 h5 break waiting on the h-file. cost objectively a full pawn, and that pawn belonged to White.

? did not take them. The move mirrors the black knight: it clears f4 so that Nc7 can go back to d5, and puts the starting position of the sequence back on the board. Sixteen seconds of thinking time stand next to it in the notation. Two moves later the same pieces stood on the same squares, Black played for the second time, and ? repeated the position for the third time, which ended the game. Both times and were the best move, both times with an evaluation that returned the advantage; both times there was no clock left to see it.

When the opponent hands you a chance, change the plan. The move you have just been calculating is no longer the same move: there is a different position on the board. A knight move that mirrors is the fastest road to a draw by repetition, and in a position you want to win it is never the answer.

IM Bakhrom Bakhrillaev (UZB, 2412)

GM Nico Georgiadis (SUI, 2535)

After 37...Nc7: the knight leaves d5 voluntarily. White to move, with one minute and nineteen seconds.

Verdict

Half a point with White against an International Master rated 123 points below him, on board two on a day where White is meant to bring the point. The first thirty-four moves were the answer you want to see on board two: a quiet build-up, the bishop for the knight at the right moment, a kingside push that freezes the black structure, and by move thirty-four an evaluation of about a pawn and a half. The last six were the answer you do not want to see. As in round six, it was the clock that decided: under five minutes from move twenty-nine, under one minute from move thirty-five, and in forty-eight seconds is the move that gave the game away without a calculation error being needed. Afterwards Black handed the chance back twice, and both times there was no clock left to accept it. Half a point, a team that needed more that day, and the same lesson repeating itself this week.

Moves

½-½

Was this analysis useful?

BognerChess app · Waiting list

Analysis like this, for your own games

The AI coach in the new BognerChess app finds the critical moments of your games and explains them in plain words. Join the waiting list and be the first to try it.

Join the waiting list

Every board of round 8

Switzerland fielded four players. Each game is annotated on its own page.